Song, Z; Florez-Perez, L; (2022) End-to-end GRU model for construction crew management. In: Teizer, J and Schultz, CPL, (eds.) Proceedings of the 29th EG-ICE International Workshop on Intelligent Computing in Engineering. (pp. pp. 196-205). European Group for Intelligent Computing in Engineering (EG-ICE): Aarhus, Denmark.

Preview

Text Final paper-EGICE-21042022-ZSLFP.pdf - Accepted Version Download (661kB) | Preview

Abstract

Crew management is critical towards improving construction task productivity. Traditional methods for crew management on-site are heavily dependent on the experience of site managers. This paper proposes an end-to-end Gated Recurrent Units (GRU) based framework which provides site managers a more reliable and robust method for managing crews and improving productivity. The proposed framework predicts task productivity of all possible crew combinations, within a given size, from the pool of available workers using an advanced GRU model. The model has been trained with an existing database of masonry work and was found to outperform other machine learning models. The results of the framework suggest which crew combinations have the highest predicted productivity and can be used by superintendents and project managers to improve construction task productivity and better plan future projects.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item